- Email: [email protected]
Stability of lanthanum-saturated montmorillonite under high pressure and high temperature conditions
Applied Clay Science 102 (2014) 51–59
Contents lists available at ScienceDirect
Applied Clay Science journal homepage: www.elsevier.com/locate/clay
Research Paper
Stability of lanthanum-saturated montmorillonite under high pressure and high temperature conditions Vicente Fiorini Stefani a,⁎, Rommulo Vieira Conceição a,b, Larissa Colombo Carniel b, Naira Maria Balzaretti a,c a b c
Programa de pós-Graduação em Ciências dos Materiais, UFRGS, Av. Bento Gonçalves, 9500, P.O. Box 15051, CEP: 91501-970 Porto Alegre, Brazil Instituto de Geociências, UFRGS, Av. Bento Gonçalves, 9500, prédio 43126, P.O. Box 15001, CEP: 91501-970 Porto Alegre, Brazil Instituto de Física, UFRGS, Av. Bento Gonçalves, 9500, prédio 43133, CEP: 91501-970 Porto Alegre, Brazil
a r t i c l e
i n f o
Article history: Received 15 December 2013 Received in revised form 16 October 2014 Accepted 18 October 2014 Available online xxxx Keywords: Smectite Interlayer water Ion exchange Radioactive waste disposal High pressure Subduction zone
a b s t r a c t Smectite has been used to capture radioactive cations through adsorption in deep radioactive waste repositories in various parts of the world. Smectite is also important in the transport of water and some trace element cations such as rare earth elements (REE), which are captured in its structure, back to the mantle in subduction environments. Such captures are based on the ionic strength of the surrounding solution and the adsorption coefficient of smectite. However, captured cations can be released from the smectite structure once the ionic strength of the solution changes. In this work, the stability of a particular smectite (montmorillonite) structure saturated with lanthanum was verified at high pressures (up to 12 GPa) and room temperature and at high pressure and high temperature (HPHT) concomitantly. La3+-montmorillonite remains stable up to 12 GPa at room temperature with a small variance in its vibrational mode. At HPHT, however, the structure becomes muscovite-like and rich in La3+. When in contact with a Ca2+-enriched solution, La3+ is partially replaced by Ca2+ in the new phase, returning to its original Ca2+-montmorillonite phase, whereas another part remains La-muscovite-like. These results were confirmed by X-ray diffraction and scanning electron microscopy with energy dispersive X-ray spectroscopy. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Smectites are clay minerals with a tetrahedral:octahedral structural ratio of 2:1. Due to isomorphic substitutions in the octahedral and tetrahedral sheets, a net negative charge is established in the structure; this charge is balanced through cation adsorption, usually in the interlayer spaces, which results in a high cation exchange capacity (CEC) (Bergaya et al., 2006). In addition to the adsorption of mono- and divalent cations such as Na+, K+, and Ca2+, smectite can also adsorb trivalent elements (e.g., trivalent rare earth elements (REE3+) and trivalent actinides), as has been shown in several studies (Takahashi et al., 1998; Stumpf et al., 2001; Bradbury and Baeysns, 2002; Coppin et al., 2002; Stumpf et al., 2002; Coppin et al., 2003; Stumpf et al., 2004; Rabung et al., 2005; Brandt et al., 2007; Tan et al., 2010). Due to these and other features, smectites are used as geochemical barriers in different contexts including secondary barriers for deep nuclear waste disposal (Pusch, 1998), where smectites are employed to adsorb actinide trivalent cations. In the geologic environment, smectites can play an important role in oceanic subduction-related zones because they transport water in higher amounts compared to micas or kaolinites; thus, some elements such as K+ and Na+ along with very minor ⁎ Corresponding author. Tel.: +1 647 713 3358. E-mail address: [email protected] (V.F. Stefani).
http://dx.doi.org/10.1016/j.clay.2014.10.012 0169-1317/© 2014 Elsevier B.V. All rights reserved.
quantities of incompatible elements such as REE3+ are transported back to the mantle, causing mantle refertilization. The different reaction mechanisms between clay minerals and cations are as follows: outer-sphere interaction, inner-sphere interaction, cation exchange within the interlayer spaces, and structure incorporation in the octahedral layer (Takahashi et al., 1998; Strawn and Sparks, 1999; Stumpf et al., 2001; Bradbury and Baeysns, 2002; Coppin et al., 2002; Stumpf et al., 2002; Coppin et al., 2003; Stumpf et al., 2004; Rabung et al., 2005; Brandt et al., 2007; Tan et al., 2010). The different modes of clay mineral–cation interaction are highly dependent on the pH of the surrounding environment. At pH values equal to or lower than 5, the adsorption is outer-sphere; while at pH values higher than 5, adsorption becomes inner-sphere (Strawn and Sparks, 1999; Stumpf et al., 2001, 2002). Several studies have shown that smectite stability is strongly dependent on temperature. Temperatures between 105 °C and 240 °C destabilize 40% of smectite, transforming it into illite in 3.4 million years (Kamel et al., 1990). At temperatures lower than 100 °C, the rate of transformation is 0.3% per million years (Chapman et al., 1984), while smectite remains unchanged for over 106 years at temperatures lower than 90 °C (Pusch and Karnland, 1988; Pusch et al., 1989). The effect of temperature on the structure of La3+-montmorillonite (La 3+-Mt) was studied by Mozas et al. (1980), who showed that La 3+-Mt loses all its interlayer water at 320 °C. In this condition,
52
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
La3+-Mt can be re-hydrated once temperature decreases. However, at temperatures higher than 500 °C, the re-hydration of La3+-Mt no longer occurs (Alba et al., 1997). Although some studies have investigated the temperature dependence of montmorillonite stability, few works have explored montmorillonite stability under high pressure conditions. When Ca2+montmorillonite (Ca2+-Mt) is exposed to pressures up to 13 GPa at room temperature, the vibrational mode of the tetrahedral Si\O bond is affected without compromising the Ca2+-Mt structure. Once pressure is released, Ca2+-Mt returns to its original structural condition (Alabarse et al., 2011). Under high pressure and high temperature (HPHT) conditions such as 7.7 GPa and 250 °C, Ca2+-Mt remains stable (Alabarse, 2009). The goal of this work is to study the effect of pressure and temperature on the stability of La3+-Mt. The results of this study will be used to understand the transportation of elements in subduction environments. In addition, we intend to transform La3+-Mt into a new in La3+-rich structure under HPHT and verify whether this new structure releases La3+ when in contact with an aqueous solution enriched in other elements. Lanthanum is analogous to actinide radionuclides due to their similar chemical properties (Krauskopf, 1986), and this experiment focuses on radioactive waste disposal. Future works will examine smectites saturated with different lanthanides (samarium, neodymium, gadolinium and lutetium).
provided elsewhere (Khvostantsev, 1974; Sherman and Stadtmuller, 1987; Stefani, 2012). This apparatus can reach pressures of up to 7.7 GPa and temperatures up to 2000 °C. The pressure cell consists of a graphite heater (height of 12.0 mm, diameter of 8.0 mm, and wall thickness of 2.0 mm) and two small disks of pyrophyllite calcinated at 1000 °C (diameter of 4.0 mm and height of 1.5 mm). To apply isostatic pressure, the La3+-Mt sample was placed inside a hexagonal boron nitride (hBN) capsule (2.0-mm internal diameter and height). Finally, the hBN capsule containing the sample was placed in between the two pyrophyllite disks. In all experiments, pressure was initially applied at room temperature and kept at the desired value for 15 min for pressure stabilization. The pressure calibration was performed using bismuth (Bi), which has phase transitions at 2.5 and 7.7 GPa (Sherman and Stadtmuller, 1987). Heating was applied simultaneously to pressure by passing an electric current through the reaction cell, which heats according to Joule's first law. Temperature calibration was carried out with a platinum and platinum rhodium thermocouple (13% Type-R). The ratio of applied voltage and temperature is known from the literature (Bundy, 1988). After the experimental run, the samples were ground in an agate mortar just before XRD analyses. The experiments performed in this apparatus have pressure and temperature accuracies of ±0.5 GPa and ±25 °C, respectively (Alabarse, 2009).
2. Experimental methods
2.3.2. Diamond anvil cell experiments A Piermari-Block diamond anvil cell (DAC) (Piermarini and Block, 1975) was used to reach pressures of up to 12 GPa at room temperature. A mixture of 1 mass% La3+-Mt powder in KBr and a small ruby were placed in a 250-μm diameter hole drilled in a Waspaloy gasket preindented to a thickness of 80 μm (Piermarini et al., 1975). Pressure was determined using the ruby technique (Piermarini et al., 1975) with an accuracy of ±0.2 GPa.
2.1. Starting material Ca2+-Mt was extracted from bentonite collected on the border of Brazil and Uruguay in the regions of Aceguá (Brazil) and Melo (Uruguay). The structural formula of the montmorillonite used in this work was determined by Calarge et al. (2003) from a bulk bentonite and is given by: 3þ þ 2þ ½Si3:87 Al0:13 O10 Al1:43 Fe0:08 Mg0:53 Ti0:01 ðOHÞ2 K0:01 Ca0:2 :
ð1Þ
The material is montmorillonite with calcium as the main interlayer cation, and the calculated layer charge was 0.47 per O10(OH)2. Ca2+-Mt was separated from bentonite by particle decantation in order to concentrate particle sizes smaller than 2 μm (Day, 1965). First, the bentonite was gently ground, and 200 g of the sample was placed into 500-ml bottles (50 g of bentonite and 300 ml of distillated water per bottle) followed by shaking for 24 h. Subsequently, the solution was left in a beaker for 24 h and 30 min to allow particles smaller than 2 μm to separate according to Stoke's law. 2.2. Cation exchange process Cation exchange was performed in order to produce La3+-Mt from Ca -Mt via the substitution of Ca2+ with La3+ using the saturation method. Ca2+-Mt (0.1 g) was added to 25 ml of 1 mol/l LaCl3 solution over 1 h. Subsequently, the solution was replaced with another 25 ml of 1 mol/l LaCl3 solution and left for 24 h. The sample was then centrifuged and washed with ethyl alcohol until the AgCl test was negative for the presence of Cl in the final material. The pH was buffered at 5 during the entire procedure. The sample was dried at room temperature. Finally, the process was repeated in order to obtain a greater amount of sample. 2+
2.3. High pressure and high temperature (HPHT) experiments 2.3.1. Toroidal chamber experiments HPHT experiments were performed on a hydraulic press with a toroidal chamber. A detailed description of the equipment used is
2.4. Analytical techniques The samples were analyzed by scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDS), X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR). SEM-EDS was performed on a sample of natural montmorillonite (Ca2+-Mt), La3+-Mt and La3+-Mt processed at 2.5 GPa and 700 °C for 8 h. The equipment used was a JEOL JSM 5800 with an acceleration voltage of 10 kV, which allows the detection of any element heavier than boron. XRD analyses were conducted using a Siemens D500 XRD powder diffractometer equipped with a CuKα source and a graphite monochromatic in the secondary beam. The spectra of all samples were obtained from 3° to 65° with a step of 0.05° at 2 s/step and from 58° and 65° with a step of 0.02° at 4 s/step in order to investigate the b-parameter. FTIR spectroscopy was performed on Ca2+-Mt and La3+-Mt samples. An in situ FTIR analysis was performed on La3+-Mt under pressure in the DAC using a Bomem FTIR model MB100 equipped with a DTGS detector and a KBr beam splitter. The spectral range was 350 to 4000 cm−1, and a total of 512 scans was performed with 4 cm−1 resolution. 3. Results and discussion 3.1. Ca2+–La3+ exchange in montmorillonite The compositions of the natural Ca2+-Mt and La3+-Mt obtained by SEM-EDS are shown in Table 1. The amount of calcium in Ca2+-Mt is in agreement with the literature (Alabarse et al., 2011). The XRD analysis (Fig. 1) of Ca2+-Mt suggested the presence of a minor amount of quartz, and Ca2+-Mt is characterized by the presence of the (001) basal plane at a distance of 15.21 Å. The 001 basal plane changes to
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59 Table 1 Average composition in mass% determined by SEM-EDS in a selected area of the natural montmorillonite (Ca2+-Mt) and montmorillonite saturated with La (La3+-Mt) samples. The total amount sums to 100%.
2+
Ca -Mt La3+-Mt
Na2O
MgO
Al2O3
SiO2
K2O
CaO
Fe2O3
LaO3
Cl
0.77 0.48
5.46 5.61
22.84 21.52
66.3 54.26
0.00 0.14
3.10 0.10
1.52 3.35
0.00 14.54
0.00 0.48
16.83 Å when the sample is saturated with ethylene glycol and 9.47 Å when it is calcined at 550 °C. After La3+ ion exchange, the amount of Ca2+ in La3+-Mt is highly reduced, while the amount of La3+ is greatly increased. Furthermore, the minor presence of Cl− indicated by the analyses suggests that La no longer exists in the chloride form. Therefore, these results indicate that the Ca2+ in the interlayer spaces of montmorillonite was exchanged by La3+. Indeed, La3+ is the only cation in our system able to replace the Ca2+ in the interlayer spaces. The structural formula calculated for La3+-Mt based on the EDS results (Table 1) is given by: 3þ 3þ ½Si3:46 Al0:54 O10 Al1:08 Fe0:16 Mg0:53 ðOHÞ2 Na0:06 La0:29 :
ð2Þ
EDS only gives a qualitative estimate of the composition; therefore, the formula is approximate. Fig. 1 shows the XRD patterns of Ca2+-Mt and La3+-Mt, with the detail of the b-parameter (reflection 060). The two patterns have very similar reflections, with a slight variance in the 001 reflection (15.23 Å for Ca2+-Mt vs. 15.80 Å for La3+-Mt). The b-parameter (in detail) indicates that both the Ca2+-Mt and La3+-Mt structures are dioctahedral. The difference in the basal distance might be associated with the presence of each cation (lanthanum and calcium) and the arrangements of their hydration shells.
53
Previous studies on Ca2+-Mt have shown that the presence of two water layers leads to a basal distance of approximately 15.1 Å (Ferrage et al., 2005), while La3+-Mt with two water layers is expected to have a basal distance of 15.8 Å (Mozas et al. 1980). The presence of water is evidenced by the peak at 1645 cm−1 in the FTIR spectrum (Fig. 2), which was measured by diffuse reflectance. The vibrational mode of La3+-Mt is very similar to that of the original 2+ Ca -Mt (Fig. 2). The band between 3000 and 3600 cm−1 corresponds to the hydroxyl groups in the octahedral and interlayer water of both Ca2+- and La3+-montmorillonite (Mookherjee and Redfern, 2002; Aranha, 2007). Peaks at 848 and 916 cm−1 due to Al\OH\Al bonding, characteristic of montmorillonite with a high amount of Al (Wagner et al., 1994), are present in both the Ca2+- and La3+-Mt spectra. The peak at 1114 cm− 1 corresponds to the Si\O bond, and the peak at 670 cm− 1 corresponds to the Si\O\Mg bond (Calvert and Prost, 1971). The band at approximately 2350 cm−1 is due to carbon dioxide adsorbed in the sample.
3.2. La3+-montmorillonite under high pressure In situ FTIR analysis was performed on the La3+-Mt sample under high pressure (up to 12 GPa) using the DAC system. The sample was dispersed in KBr (1 mass% of La3+-Mt, 99 mass% of KBr) and measured by transmittance at different pressures. The spectra obtained at all pressures were very similar, with the exception of the peak at 1040 cm− 1 (Fig. 3), which corresponds to the tetrahedral Si\O bond (Mookherjee and Redfern, 2002); this peak shifted from 1039 cm− 1 at atmospheric pressure to 1063 cm − 1 at 12 GPa. The change in this bond frequency from lower to higher frequency with increasing pressure was also observed by Alabarse et al. (2011) for Ca 2+-Mt and is related to the approach of the Si\O apical bond from the tetrahedron, which increases its vibrational modes. This process is reversible; as the pressure is released, the peak returns to its original position (Fig. 3).
Fig. 1. XRD patterns of La3+-Mt and natural Ca3+-Mt with the respective planes and distances. Both patterns indicate minor amounts of quartz (Q). The figure in detail shows the bparameters, indicating dioctahedral structures for both samples.
54
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
Fig. 2. FTIR spectrum of La3+-Mt and natural Ca2+-Mt.
3.3. La3+-montmorillonite under high pressure and high temperature (HPHT) La3+-Mt was processed at a pressure of 2.5 GPa and temperatures of 200 °C, 250 °C, 300 °C, 400 °C, 500 °C, 650 °C and 700 °C as well as at a pressure of 7.7 GPa and temperatures of 250 °C, 300 °C, 350 °C and 900 °C. EDS analysis was performed for the experiment carried out at 2.5 GPa and 700 °C for 8 h on the La3+-Mt sample. The La3+ is still
Fig. 3. In situ FTIR spectra for La3+-Mt under high pressures using the DAC system. The up arrows indicate that the pressure is being applied, while the down arrows indicate that the pressure is being released.
present in the structure at HPHT, as indicated by the high amount measured by EDS (13.20 mass%); the approximate structural formula is given by: 3þ 3þ ½Si3:13 Al0:87 O10 Al1:07 Fe0:13 Mg0:83 ðOHÞ2 La0:28 :
ð3Þ
XRD analyses were performed on all samples. Fig. 4 shows the XRD results for experiments performed at 2.5 GPa, and Fig. 5 shows the patterns for experiments performed at 7.7 GPa. All experiments were carried out for 8 h. At 2.5 GPa, the La3+-Mt structure remains stable up to 200 °C. At 250 °C, however, the (001) basal reflection shifts to a higher angle, corresponding to a basal distance of 10.5 Å; together with the other reflections, this infers that the structure has collapsed and changed to a La3+-muscovite-like structure. Although an illite-like structure was expected, the XRD reflections are in better agreement with a muscovite-like structure. For this reason, we refer to this new HPHT phase as La3+-muscovite-like from now on. Based on the structural formula calculated from the EDS results (Eq. (3)), the structure has a high layer charge; however, the nature of this phenomenon cannot be determined. As temperature increases, the La3+-muscovite-like diffraction reflections become narrower and more intense due to the more ordered structure. At 7.7 GPa, the structure of La3+-Mt is stable up to 300 °C and turns to La3+-muscovite-like at 350 °C. At 900 °C, the quartz present in the starting material becomes coesite, and the La3+-muscovite-like structure remains stable. In all experiments at both 2.5 and 7.7 GPa, the La3+-muscovite-like structure remains dioctahedral, and La3+ is part of the muscovite-like structure. When the La3+-Mt structure becomes La3+-muscovite-like, the structure loses its interlayer water, and the distance between layers drops to approximately 10.5 Å. Fig. 6 shows the detailed changes observed in the 001 reflection of the La-muscovite-like structure along with the corresponding basal distances for each HPHT condition. As
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
55
Fig. 4. XRD patterns of La3+-Mt at 2.5 GPa and different temperatures. The 001 peak of muscovite, indicated by an asterisk (*), is better specified in Fig. 7. At 200 °C, the structure remains as montmorillonite, while the patterns show muscovite-like structures at temperatures equal to and higher than 250 °C. The patterns also indicate the presence of quartz and contamination due to hBN.
Fig. 5. XRD patterns of La3+-Mt at 7.7 GPa and different temperatures. The 001 peak of La-muscovite, denoted by an asterisk (*), is better specified in Fig. 7. The structure is La3+-Mt up to a temperature of 300 °C, while it becomes La3+-muscovite-like at 350 °C and higher. The patterns also show reflections of quartz and coesite as well as low intensity reflections of hBN for the high pressure assemblage.
56
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
Fig. 6. The detailed 001 reflection of La3+-muscovite-like samples at different pressure and temperature conditions.
temperature increases, the structure becomes better ordered and the 001 reflections shift to higher angles. Ethylene glycol (EG) saturation tests were performed on the La3+muscovite-like samples obtained at 2.5 GPa/250 °C and 2.5 GPa/ 650 °C (Fig. 7) in order to analyze the behavior of their structures under these conditions. These samples were selected for EG testing because the 2.5 GPa/250 °C sample is one that is most expected to resaturate due to the lower temperature condition and the 001 reflection shape (see Fig. 6); the sample obtained at 2.5 GPa/650 °C was selected because its conditions are closest to those of the sample used in the Ca2+ adsorption experiment of the La3+-muscovite-like structure (see Section 4). None of these samples showed total or partial expansion of
the 001 distance, suggesting that there is no illite–smectite structure in these samples. However, the simple EG treatment is not sufficient to re-expand all smectite layers (Ferrage et al., 2011). EG saturation was performed as described in Bradley (1945). Fig. 8 summarizes the pressure and temperature conditions at which La3+-Mt collapses irreversibly (became a La3+-muscovite-like structure) and reversibly (remains as La3+-Mt) after the experiments. In the figure, black squares represent the experimental conditions at which the La3+-Mt structure returns to its original phase. In contrast, the black triangles represent the experimental conditions at which the La3+-Mt structure collapses to become La3+-muscovite-like and does not rehydrate afterwards. When the pressure increases, the interlayer water does leave the structure due to osmotic pressure; therefore, the structure returns to its original phase after the release of pressure. On the other hand, as temperature increases, the interlayer water has more mobility and can dehydrate, leading to an irreversible phase transition. Yet, as pressure increases, higher temperatures are necessary to generate an irreversible phase transition (Fig. 8). At these conditions, we assume that the interlayer spaces collapse and trap the lanthanum ions; we call this phase La3+-muscovite like. The phase diagram shown in Fig. 8 is similar to that of water (Wagner et al., 1994; Saitta and Datchi, 2003; Lin et al., 2004; Vega et al., 2009). To verify the stability of La inside the La3+-muscovite-like structure, we performed an ion exchange experiment in which we tried to replace La with Ca2+ in the La3+-muscovite-like structure. The experimental sample was the one produced at 2.5 GPa/700 °C/8 h because its structure exhibits good crystallinity. An XRD scan between 3° and 10° with a step of 0.02° at 4 s/step was carried out after the ion exchange experiment in order to verify the position of the first reflection (001). The diffraction pattern has two reflections: 5.73° (d(001) = 15.25 Å) and 8.62° (d(001) = 10.24 Å; Fig. 9). Therefore, two different structures are present: one that remains unchanged (La3+-muscovite-like) and another that changes into Ca 2+-Mt (15.25 Å). This result implies the presence of a mixed layer structure due to the partial readsorption of Ca 2+ by the La 3+-muscovite-like structure, turning it into the original
Fig 7. Comparison of La3+-Mt obtained at 2.5 GPa/250 °C and 2.5 GPa/650 °C saturated with ethylene glycol.
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
57
Fig. 8. Diagram based on the obtained results. Squares represent the conditions at which the structure returns to the original phase, while triangles represent the conditions at which the structure transforms.
Ca 2+ -Mt phase and releasing La3+ to the solution. As the cation exchange was performed at a pH of approximately 5, we presume that this pH is responsible for the partial cation leaching of the HPHT La3+-muscovite-like structure. Fig. 11 shows the backscattering image of this sample (Fig. 10a) along with the distributions of Ca2+ and La3+ (Fig. 10b and c, respectively). Table 2 confirms the
presence of areas enriched in Ca2+ and depleted in La3+ along with other areas enriched in La3+ and depleted in Ca2+. The approximate structural formulas calculated from the data in Table 2 are given in Eq. (4) for area 1 and Eq. (5) for area 2. 3þ 2þ 3þ Si4 O10 Al0:88 Fe0:03 Mg0:05 ðOHÞ2 Na0:65 Ca0:95 La0:08 ð4Þ
Fig. 9. XRD pattern of the La3+-muscovite-like structure processed at 2.5 GPa/700 °C/8 h and subsequently saturated with calcium. The reflection at 15.25 Å corresponds to La3+-Mt, while the reflection at 10.24 Å corresponds to the La3+-muscovite-like structure.
58
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
Fig. 10. a) Backscattering image of the La3+-muscovite-like structure saturated with Ca2+ from the areas (1 and 2) where EDS was performed, b) Ca2+ distribution and c) La3+ distribution.
3þ 2þ 3þ ½Si3:15 Al0:85 O10 Al0:87 Fe0:249 Mg0:41 ðOHÞ2 Ca0:02 La0:43 :
ð5Þ
4. Significance to geologic and radioactive waste deposits Our experiments are applicable to two scenarios: subduction environments and radioactive waste disposal. Some amount of REE can be found in authigenic illite–smectite, as was shown by Uysal and Golding (2003); greater enrichment of heavy REE3+ is associated with a high degree of illitization, whereas less illitic clays show a higher concentration of light REE3+. During the deep burial of sediments in the subduction zone, hydrous minerals such as micas are considered to be the major carriers of near-surface H2O that is able to enter in the lower crust and mantle (Huang and Wyllie, 1973). However, smectites contain a much larger amount of water (± 15 mass%) than micas (±4.7 mass%) or even kaolinites (±13.8 mass%) and are widespread in pelagic sediments. La3+-montmorillonite (or La3+-muscovite or La3+-illite) does not occur in nature, although minor amounts of La3+ can be trapped in the interlayer spaces of montmorillonite, as described elsewhere, to share the interlayer space with other major cations such as K+, Ca2+ or Na+ (Huang and Wyllie, 1973; Uysal and Golding, 2003). When the mineral reaches pressure and temperature conditions that lead to dehydration according to the diagram above (Fig. 8), the water lost from the mineral can decrease the melting point of the surrounding rocks and promote their partial melting. Our experiments of Ca2+–La3+ replacement in the La3+-muscovite-like structure also suggest that in the presence of fluids, La3+ can be partially released from the La3+-muscovite-like structure to the surroundings, enriching the mantle in water and incompatible elements (e.g., La3+), possibly causing mantle metasomatism. However, part of the La3+ may still remain in the La3+-muscovite-like structure in deeper and hotter regions; thus, mantle incompatible element enrichment can be extended to higher pressure conditions. For radioactive waste disposal, our experiments show that La3+ (or its actinide counterpart) can be easily placed inside of La3+-Mt via cation exchange. If this mineral is exposed to high pressures and temperatures (beyond 2.5 GPa and 250 °C), it transforms into a La3+muscovite-like structure by trapping La3+. These results suggest a different approach to radioactive waste disposal where the hazardous elements are trapped in a stable crystalline structure. However, such Table 2 EDS in mass% from different areas (1 and 2) of La-muscovite saturated with Ca. The total amount sums to 100%.
Area 1 Area 2
Na2O
MgO
Al2O3
SiO2
Cl
CaO
Fe2O3
LaO3
5.30 0.00
0.57 4.17
11.75 22.16
63.56 47.74
0.00 0.04
13.96 0.33
0.68 5.02
4.18 20.54
trapping has been shown to be inefficient or partially inefficient as La3+ can be totally or partially released from La3+-Mt and La3+-muscovite-like structures, respectively, if the surrounding water leaches these mineral structures. Therefore, further studies are needed to identify a structure that completely traps lanthanum cations regardless of the surrounding environment. One of these high pressure structures could be garnet, which has a greater ability to contain REE, particularly heavy-REE3+, which better mimic the actinides. Our group is currently developing studies on this subject. 5. Conclusion Ion exchange with La3+ was achieved in Ca2+-Mt, and the structure remained very similar with the exception of the cations in the interlayer spaces. High pressures applied to the La3+-Mt structure at room temperature resulted in minor variations in the vibrational mode of the tetrahedral Si\O bond; however, this effect was reversible. When pressure and temperature were simultaneously applied to La3+-Mt, its structure remained stable up to around 2.5 GPa/250 °C and 7.7 GPa/ 350 °C; in these cases, the interlayer water had enough mobility to leave the structure, transforming La3+-Mt into a La3+-muscovite-like structure. At this point, no evidence for an illite–smectite structure was observed in our experiments. With increasing temperature, the structure became more ordered, and the HLW radionuclides could be trapped into the structure before being deposited in a multi-barrier system. Nevertheless, the La3+-muscovite-like structure in contact with an aqueous solution rich in other cations (Ca2+, for instance) would cause partial La3+ leaching; in this case, part of the structure would transform to Mt, while part of the structure would remain with the cation trapped in a La3+-muscovite-like structure. Acknowledgments We would like to thank the financial support provided by Comissão Nacional de Energia Nuclear (CNEN) from Brazil. References Alabarse, F.G., 2009. Análise da estabilidade estrutural da esmectita sob altas pressões e altas temperaturasMaster Thesis UFRGS, Porto Alegre, Brazil. Alabarse, F.G., Conceicao, R.V., Balzaretti, N.M., Schenato, F., Xavier, A.M., 2011. In-situ FTIR analyses of bentonite under high-pressure. Appl. Clay Sci. 51, 202–208. Alba, M.D., Alvero, R., Becerro, A.I., Castro, M.A., Muñoz-Paez, A., Trillo, J.M., 1997. Study of the reversibility on the local La3+ environment after thermal and drying treatments in lanthanum-exchanged smectites. Nucl. Inst. Methods Phys. Res. B 133, 34–38. Aranha, I.B., 2007. Preparação, caracterização e propriedades de argilas organofílicas. Doctorate dissertation, UFRGS, Porto Alegre, Brazil. Bergaya, F., Theng, B.K.G., Lagaly, G., 2006. Handbook of clay science. Developments in Clay Science vol. 1. Elsevier Science, Amsterdam. Bradbury, M.H., Baeysns, B., 2002. Sorption of Eu on Na- and Ca-montmorillonites: experimental investigations and modeling with cations exchange and surface complexation. Geochim. Cosmochim. Acta 66, 2325–2334. Bradley, W.F., 1945. Molecular association between montmorillonite and some polyfunctional organic liquids. J. Am. Chem. Soc. 67, 975–981.
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59 Brandt, H., Bosbach, D., Panak, P.J., Fanghanel, T., 2007. Structural incorporation of Cm(III) in trioctahedral smectite hectorite: a time-resolved laser fluorescence spectroscopy (TRLFS) study. Geochim. Cosmochim. Acta 71, 145–154. Bundy, F.P., 1988. Ultra-high pressure apparatus. Phys. Rep. 167, 133–176. Calarge, L., Meunier, A., Formoso, M.L., 2003. A bentonite bed in Aceguá (RS, Brasil) and Melo (Uruguay) areas: a highly crystallized montmorillonite. J. S. Am. Earth Sci. 16, 187–198. Calvert, R., Prost, R., 1971. Cation migration into empty octahedral sites and surface properties of clays. Clay Clay Miner. 19, 175–186. Chapman, N.A., McKanley, I.G., Smelhe, J.A.T., 1984. The potential of natural analogues in assessing systems for deep disposal of high-level radioactive waste. Nagra Technical Report NTB, Baden, 84-41. Coppin, F., Berger, G., Bauer, A., Castet, S., Loubet, M., 2002. Sorption of lanthanides on smectite and kaolinite. Chem. Geol. 182, 57–68. Coppin, F., Castet, S., Berger, G., Loubet, M., 2003. Microscopy reversibility of Sm and Yb sorption onto smectite and kaolinite: experimental evidence. Geochim. Cosmochim. Acta 67, 2515–2527. Day, P.R., 1965. Particle fractionation and particle size analysis. In: Black, C.A. (Ed.), Methods of Soil Analysis. American Society of Agronomy, pp. 545–567. Ferrage, E., Lanson, B., Sakharov, B.A., Drits, V.A., 2005. Investigation of smectite hydration properties by modeling experimental X-ray diffraction patterns: Part I. Montmorillonite hydration properties. Am. Mineral. 90, 1358–1374. Ferrage, E., Vidal, O., Mosser-Ruck, R., Cathelineau, M., Cuadros, J., 2011. A reinvestigation of smectite illitization in experimental hydrothermal conditions: results from X-ray diffraction and transmission electron microscopy. Am. Mineral. 96, 207–223. Huang, W.L., Wyllie, P.J., 1973. Muscovite dehydration and melting in the crust and subduction oceanic sediments. Earth Planet. Sci. Lett. 18, 133–136. Kamel, G., Arai, T., Yusa, Y., Sasaki, N., Sakuramoto, Y., 1990. Estimation of illitization rate of smectite from the thermal history of Murakami Deposit, Japan. Mater. Res. Soc. Symp. Proc. 176, 657–663. Khvostantsev, L.G., 1974. High temperatures–high pressures 16, p. 165. Krauskopf, K.B., 1986. Thorium and rare-earth metals as analogs for actinide elements. Chem. Geol. 55, 323–335. Lin, J., Militzer, B., Struzhkin, V.V., Gregoryanz, E., Hemley, R.J., Mao, H., 2004. High pressure–temperature Raman measurements of H2O melting to 22 GPa and 900 K. J. Chem. Phys. 121, 8423–8427. Mookherjee, M., Redfern, S.A.T., 2002. A high-temperature Fourier transform infrared study of the interlayer and Si\O-stretching region in phengite-2M1. Clay Miner. 37, 323–336. Mozas, T., Bruque, S., Rodriguez, A., 1980. Effect of thermal treatment on lanthanide montmorillonites: dehydration. Clay Miner. 15, 421428. Piermarini, G.J., Block, S., 1975. Rev. Sci. Instrum. 46, 973. Piermarini, G.J., Block, S., Barnett, J.D., Forman, R.A., 1975. Calibration of the pressure dependence of the R1 ruby fluorescence line to 195 kbar. J. Appl. Phys. 46, 2774–2780. Pusch, R., 1998. Transport of radionuclides in smectite clay. In: Parker, A., Rae, J.E. (Eds.), Environmental Interactions of Clays — Clays and Environment. Springer.
59
Pusch, R., Karnland, O., 1988. Hydrothermal effects on montmorillonite: a preliminary study. SKB Technical Report, Stockholm 88-15. Pusch, R., Karland, O., Muurmen, A., 1989. Transport and microstructural phenomena in bentonite clay with respect to the behavior and influence of Na, Cu, and U. SKB Technical Report, Stockholm 89-34. Rabung, T.H., Pierret, M.C., Bauer, A., Geckeis, H., Bradbury, M.H., Baeysns, B., 2005. Sorption of Eu(III)/Cm(III) on Ca-montmorillonite and Na-illite. Part 1: batch sorption and time-resolved laser fluorescence spectroscopy experiments. Geochim. Cosmochim. Acta 69, 5393–5402. Saitta, A.M., Datchi, F., 2003. Structure and phase diagram of high-density water: the role of interstitial molecules. Phys. Rev. 67, 020201(R). Sherman, W.F., Stadtmuller, A.A., 1987. Experimental Techniques in High Pressure Research. John Wiley & Sons Ltd., London. Stefani, V.F., 2012. Estabilidade estrutural da esmectita dopada com lantânio sob altas pressões e altas temperaturasMaster Thesis UFRGS, Porto Alegre, Brasil. Strawn, D.G., Sparks, D.L., 1999. The use of XAFS to distinguish between inner- and outersphere lead adsorption complexes on montmorillonite. J. Colloid Interface Sci. 216, 257–269. Stumpf, T., Bauer, A., Coppin, F., Kim, J.I., 2001. Time-resolved laser fluorescence spectroscopy study of the sorption of Cm(III) onto smectite and kaolinite. Environ. Sci. Technol. 35, 3691–3694. Stumpf, T., Bauer, A., Coppin, F., Fanghanel, Th., Kim, J.I., 2002. Inner-sphere, outer-sphere and ternary surface complexes: a TRLFS study of the sorption process of Eu(III) onto smectite and kaolinite. Radiochim. Acta 90, 345–349. Stumpf, T., Hennig, C., Bauer, A., Denecke, M.A., Fanghanel, T., 2004. An EXFAS and TRLFS study of the sorption of trivalent actinides onto smectite and kaolinite. Radiochim. Acta 92, 133–138. Takahashi, Y., Minai, Y., Kimura, T., Tominaga, T., 1998. Adsorption of Eu(III) and Am(III) on kaolinite and montmorillonite in the presence of humic acid. J. Radioanal. Nucl. Chem. 234, 277–282. Tan, X., Fang, M., Wang, X., 2010. Sorption speciation of lanthanides/actinides on minerals by TRLFS, EXAFS and DFT studies: a review. Molecules 15, 8431–8468. Uysal, T., Golding, S.D., 2003. Rare earth element fractionation in authigenic illite– smectite from Late Permian clastic rocks, Bowen Basin, Australia: implications for physico-chemical environments of fluids during illitization. Chem. Geol. 193, 167–179. Vega, C., Abascal, J.L.F., Conde, M.M., Aragones, J.L., 2009. What ice can teach us about water interactions: a critical comparison of the performance of different water models. Faraday Discuss. 141, 251–276. Wagner, W., Saul, A., Pruβ, A., 1994. International equations for the pressure along the melting and along the sublimation curve of ordinary water substance. J. Phys. Chem. Ref. Data 23, 515–527.
Contents lists available at ScienceDirect
Applied Clay Science journal homepage: www.elsevier.com/locate/clay
Research Paper
Stability of lanthanum-saturated montmorillonite under high pressure and high temperature conditions Vicente Fiorini Stefani a,⁎, Rommulo Vieira Conceição a,b, Larissa Colombo Carniel b, Naira Maria Balzaretti a,c a b c
Programa de pós-Graduação em Ciências dos Materiais, UFRGS, Av. Bento Gonçalves, 9500, P.O. Box 15051, CEP: 91501-970 Porto Alegre, Brazil Instituto de Geociências, UFRGS, Av. Bento Gonçalves, 9500, prédio 43126, P.O. Box 15001, CEP: 91501-970 Porto Alegre, Brazil Instituto de Física, UFRGS, Av. Bento Gonçalves, 9500, prédio 43133, CEP: 91501-970 Porto Alegre, Brazil
a r t i c l e
i n f o
Article history: Received 15 December 2013 Received in revised form 16 October 2014 Accepted 18 October 2014 Available online xxxx Keywords: Smectite Interlayer water Ion exchange Radioactive waste disposal High pressure Subduction zone
a b s t r a c t Smectite has been used to capture radioactive cations through adsorption in deep radioactive waste repositories in various parts of the world. Smectite is also important in the transport of water and some trace element cations such as rare earth elements (REE), which are captured in its structure, back to the mantle in subduction environments. Such captures are based on the ionic strength of the surrounding solution and the adsorption coefficient of smectite. However, captured cations can be released from the smectite structure once the ionic strength of the solution changes. In this work, the stability of a particular smectite (montmorillonite) structure saturated with lanthanum was verified at high pressures (up to 12 GPa) and room temperature and at high pressure and high temperature (HPHT) concomitantly. La3+-montmorillonite remains stable up to 12 GPa at room temperature with a small variance in its vibrational mode. At HPHT, however, the structure becomes muscovite-like and rich in La3+. When in contact with a Ca2+-enriched solution, La3+ is partially replaced by Ca2+ in the new phase, returning to its original Ca2+-montmorillonite phase, whereas another part remains La-muscovite-like. These results were confirmed by X-ray diffraction and scanning electron microscopy with energy dispersive X-ray spectroscopy. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Smectites are clay minerals with a tetrahedral:octahedral structural ratio of 2:1. Due to isomorphic substitutions in the octahedral and tetrahedral sheets, a net negative charge is established in the structure; this charge is balanced through cation adsorption, usually in the interlayer spaces, which results in a high cation exchange capacity (CEC) (Bergaya et al., 2006). In addition to the adsorption of mono- and divalent cations such as Na+, K+, and Ca2+, smectite can also adsorb trivalent elements (e.g., trivalent rare earth elements (REE3+) and trivalent actinides), as has been shown in several studies (Takahashi et al., 1998; Stumpf et al., 2001; Bradbury and Baeysns, 2002; Coppin et al., 2002; Stumpf et al., 2002; Coppin et al., 2003; Stumpf et al., 2004; Rabung et al., 2005; Brandt et al., 2007; Tan et al., 2010). Due to these and other features, smectites are used as geochemical barriers in different contexts including secondary barriers for deep nuclear waste disposal (Pusch, 1998), where smectites are employed to adsorb actinide trivalent cations. In the geologic environment, smectites can play an important role in oceanic subduction-related zones because they transport water in higher amounts compared to micas or kaolinites; thus, some elements such as K+ and Na+ along with very minor ⁎ Corresponding author. Tel.: +1 647 713 3358. E-mail address: [email protected] (V.F. Stefani).
http://dx.doi.org/10.1016/j.clay.2014.10.012 0169-1317/© 2014 Elsevier B.V. All rights reserved.
quantities of incompatible elements such as REE3+ are transported back to the mantle, causing mantle refertilization. The different reaction mechanisms between clay minerals and cations are as follows: outer-sphere interaction, inner-sphere interaction, cation exchange within the interlayer spaces, and structure incorporation in the octahedral layer (Takahashi et al., 1998; Strawn and Sparks, 1999; Stumpf et al., 2001; Bradbury and Baeysns, 2002; Coppin et al., 2002; Stumpf et al., 2002; Coppin et al., 2003; Stumpf et al., 2004; Rabung et al., 2005; Brandt et al., 2007; Tan et al., 2010). The different modes of clay mineral–cation interaction are highly dependent on the pH of the surrounding environment. At pH values equal to or lower than 5, the adsorption is outer-sphere; while at pH values higher than 5, adsorption becomes inner-sphere (Strawn and Sparks, 1999; Stumpf et al., 2001, 2002). Several studies have shown that smectite stability is strongly dependent on temperature. Temperatures between 105 °C and 240 °C destabilize 40% of smectite, transforming it into illite in 3.4 million years (Kamel et al., 1990). At temperatures lower than 100 °C, the rate of transformation is 0.3% per million years (Chapman et al., 1984), while smectite remains unchanged for over 106 years at temperatures lower than 90 °C (Pusch and Karnland, 1988; Pusch et al., 1989). The effect of temperature on the structure of La3+-montmorillonite (La 3+-Mt) was studied by Mozas et al. (1980), who showed that La 3+-Mt loses all its interlayer water at 320 °C. In this condition,
52
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
La3+-Mt can be re-hydrated once temperature decreases. However, at temperatures higher than 500 °C, the re-hydration of La3+-Mt no longer occurs (Alba et al., 1997). Although some studies have investigated the temperature dependence of montmorillonite stability, few works have explored montmorillonite stability under high pressure conditions. When Ca2+montmorillonite (Ca2+-Mt) is exposed to pressures up to 13 GPa at room temperature, the vibrational mode of the tetrahedral Si\O bond is affected without compromising the Ca2+-Mt structure. Once pressure is released, Ca2+-Mt returns to its original structural condition (Alabarse et al., 2011). Under high pressure and high temperature (HPHT) conditions such as 7.7 GPa and 250 °C, Ca2+-Mt remains stable (Alabarse, 2009). The goal of this work is to study the effect of pressure and temperature on the stability of La3+-Mt. The results of this study will be used to understand the transportation of elements in subduction environments. In addition, we intend to transform La3+-Mt into a new in La3+-rich structure under HPHT and verify whether this new structure releases La3+ when in contact with an aqueous solution enriched in other elements. Lanthanum is analogous to actinide radionuclides due to their similar chemical properties (Krauskopf, 1986), and this experiment focuses on radioactive waste disposal. Future works will examine smectites saturated with different lanthanides (samarium, neodymium, gadolinium and lutetium).
provided elsewhere (Khvostantsev, 1974; Sherman and Stadtmuller, 1987; Stefani, 2012). This apparatus can reach pressures of up to 7.7 GPa and temperatures up to 2000 °C. The pressure cell consists of a graphite heater (height of 12.0 mm, diameter of 8.0 mm, and wall thickness of 2.0 mm) and two small disks of pyrophyllite calcinated at 1000 °C (diameter of 4.0 mm and height of 1.5 mm). To apply isostatic pressure, the La3+-Mt sample was placed inside a hexagonal boron nitride (hBN) capsule (2.0-mm internal diameter and height). Finally, the hBN capsule containing the sample was placed in between the two pyrophyllite disks. In all experiments, pressure was initially applied at room temperature and kept at the desired value for 15 min for pressure stabilization. The pressure calibration was performed using bismuth (Bi), which has phase transitions at 2.5 and 7.7 GPa (Sherman and Stadtmuller, 1987). Heating was applied simultaneously to pressure by passing an electric current through the reaction cell, which heats according to Joule's first law. Temperature calibration was carried out with a platinum and platinum rhodium thermocouple (13% Type-R). The ratio of applied voltage and temperature is known from the literature (Bundy, 1988). After the experimental run, the samples were ground in an agate mortar just before XRD analyses. The experiments performed in this apparatus have pressure and temperature accuracies of ±0.5 GPa and ±25 °C, respectively (Alabarse, 2009).
2. Experimental methods
2.3.2. Diamond anvil cell experiments A Piermari-Block diamond anvil cell (DAC) (Piermarini and Block, 1975) was used to reach pressures of up to 12 GPa at room temperature. A mixture of 1 mass% La3+-Mt powder in KBr and a small ruby were placed in a 250-μm diameter hole drilled in a Waspaloy gasket preindented to a thickness of 80 μm (Piermarini et al., 1975). Pressure was determined using the ruby technique (Piermarini et al., 1975) with an accuracy of ±0.2 GPa.
2.1. Starting material Ca2+-Mt was extracted from bentonite collected on the border of Brazil and Uruguay in the regions of Aceguá (Brazil) and Melo (Uruguay). The structural formula of the montmorillonite used in this work was determined by Calarge et al. (2003) from a bulk bentonite and is given by: 3þ þ 2þ ½Si3:87 Al0:13 O10 Al1:43 Fe0:08 Mg0:53 Ti0:01 ðOHÞ2 K0:01 Ca0:2 :
ð1Þ
The material is montmorillonite with calcium as the main interlayer cation, and the calculated layer charge was 0.47 per O10(OH)2. Ca2+-Mt was separated from bentonite by particle decantation in order to concentrate particle sizes smaller than 2 μm (Day, 1965). First, the bentonite was gently ground, and 200 g of the sample was placed into 500-ml bottles (50 g of bentonite and 300 ml of distillated water per bottle) followed by shaking for 24 h. Subsequently, the solution was left in a beaker for 24 h and 30 min to allow particles smaller than 2 μm to separate according to Stoke's law. 2.2. Cation exchange process Cation exchange was performed in order to produce La3+-Mt from Ca -Mt via the substitution of Ca2+ with La3+ using the saturation method. Ca2+-Mt (0.1 g) was added to 25 ml of 1 mol/l LaCl3 solution over 1 h. Subsequently, the solution was replaced with another 25 ml of 1 mol/l LaCl3 solution and left for 24 h. The sample was then centrifuged and washed with ethyl alcohol until the AgCl test was negative for the presence of Cl in the final material. The pH was buffered at 5 during the entire procedure. The sample was dried at room temperature. Finally, the process was repeated in order to obtain a greater amount of sample. 2+
2.3. High pressure and high temperature (HPHT) experiments 2.3.1. Toroidal chamber experiments HPHT experiments were performed on a hydraulic press with a toroidal chamber. A detailed description of the equipment used is
2.4. Analytical techniques The samples were analyzed by scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDS), X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR). SEM-EDS was performed on a sample of natural montmorillonite (Ca2+-Mt), La3+-Mt and La3+-Mt processed at 2.5 GPa and 700 °C for 8 h. The equipment used was a JEOL JSM 5800 with an acceleration voltage of 10 kV, which allows the detection of any element heavier than boron. XRD analyses were conducted using a Siemens D500 XRD powder diffractometer equipped with a CuKα source and a graphite monochromatic in the secondary beam. The spectra of all samples were obtained from 3° to 65° with a step of 0.05° at 2 s/step and from 58° and 65° with a step of 0.02° at 4 s/step in order to investigate the b-parameter. FTIR spectroscopy was performed on Ca2+-Mt and La3+-Mt samples. An in situ FTIR analysis was performed on La3+-Mt under pressure in the DAC using a Bomem FTIR model MB100 equipped with a DTGS detector and a KBr beam splitter. The spectral range was 350 to 4000 cm−1, and a total of 512 scans was performed with 4 cm−1 resolution. 3. Results and discussion 3.1. Ca2+–La3+ exchange in montmorillonite The compositions of the natural Ca2+-Mt and La3+-Mt obtained by SEM-EDS are shown in Table 1. The amount of calcium in Ca2+-Mt is in agreement with the literature (Alabarse et al., 2011). The XRD analysis (Fig. 1) of Ca2+-Mt suggested the presence of a minor amount of quartz, and Ca2+-Mt is characterized by the presence of the (001) basal plane at a distance of 15.21 Å. The 001 basal plane changes to
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59 Table 1 Average composition in mass% determined by SEM-EDS in a selected area of the natural montmorillonite (Ca2+-Mt) and montmorillonite saturated with La (La3+-Mt) samples. The total amount sums to 100%.
2+
Ca -Mt La3+-Mt
Na2O
MgO
Al2O3
SiO2
K2O
CaO
Fe2O3
LaO3
Cl
0.77 0.48
5.46 5.61
22.84 21.52
66.3 54.26
0.00 0.14
3.10 0.10
1.52 3.35
0.00 14.54
0.00 0.48
16.83 Å when the sample is saturated with ethylene glycol and 9.47 Å when it is calcined at 550 °C. After La3+ ion exchange, the amount of Ca2+ in La3+-Mt is highly reduced, while the amount of La3+ is greatly increased. Furthermore, the minor presence of Cl− indicated by the analyses suggests that La no longer exists in the chloride form. Therefore, these results indicate that the Ca2+ in the interlayer spaces of montmorillonite was exchanged by La3+. Indeed, La3+ is the only cation in our system able to replace the Ca2+ in the interlayer spaces. The structural formula calculated for La3+-Mt based on the EDS results (Table 1) is given by: 3þ 3þ ½Si3:46 Al0:54 O10 Al1:08 Fe0:16 Mg0:53 ðOHÞ2 Na0:06 La0:29 :
ð2Þ
EDS only gives a qualitative estimate of the composition; therefore, the formula is approximate. Fig. 1 shows the XRD patterns of Ca2+-Mt and La3+-Mt, with the detail of the b-parameter (reflection 060). The two patterns have very similar reflections, with a slight variance in the 001 reflection (15.23 Å for Ca2+-Mt vs. 15.80 Å for La3+-Mt). The b-parameter (in detail) indicates that both the Ca2+-Mt and La3+-Mt structures are dioctahedral. The difference in the basal distance might be associated with the presence of each cation (lanthanum and calcium) and the arrangements of their hydration shells.
53
Previous studies on Ca2+-Mt have shown that the presence of two water layers leads to a basal distance of approximately 15.1 Å (Ferrage et al., 2005), while La3+-Mt with two water layers is expected to have a basal distance of 15.8 Å (Mozas et al. 1980). The presence of water is evidenced by the peak at 1645 cm−1 in the FTIR spectrum (Fig. 2), which was measured by diffuse reflectance. The vibrational mode of La3+-Mt is very similar to that of the original 2+ Ca -Mt (Fig. 2). The band between 3000 and 3600 cm−1 corresponds to the hydroxyl groups in the octahedral and interlayer water of both Ca2+- and La3+-montmorillonite (Mookherjee and Redfern, 2002; Aranha, 2007). Peaks at 848 and 916 cm−1 due to Al\OH\Al bonding, characteristic of montmorillonite with a high amount of Al (Wagner et al., 1994), are present in both the Ca2+- and La3+-Mt spectra. The peak at 1114 cm− 1 corresponds to the Si\O bond, and the peak at 670 cm− 1 corresponds to the Si\O\Mg bond (Calvert and Prost, 1971). The band at approximately 2350 cm−1 is due to carbon dioxide adsorbed in the sample.
3.2. La3+-montmorillonite under high pressure In situ FTIR analysis was performed on the La3+-Mt sample under high pressure (up to 12 GPa) using the DAC system. The sample was dispersed in KBr (1 mass% of La3+-Mt, 99 mass% of KBr) and measured by transmittance at different pressures. The spectra obtained at all pressures were very similar, with the exception of the peak at 1040 cm− 1 (Fig. 3), which corresponds to the tetrahedral Si\O bond (Mookherjee and Redfern, 2002); this peak shifted from 1039 cm− 1 at atmospheric pressure to 1063 cm − 1 at 12 GPa. The change in this bond frequency from lower to higher frequency with increasing pressure was also observed by Alabarse et al. (2011) for Ca 2+-Mt and is related to the approach of the Si\O apical bond from the tetrahedron, which increases its vibrational modes. This process is reversible; as the pressure is released, the peak returns to its original position (Fig. 3).
Fig. 1. XRD patterns of La3+-Mt and natural Ca3+-Mt with the respective planes and distances. Both patterns indicate minor amounts of quartz (Q). The figure in detail shows the bparameters, indicating dioctahedral structures for both samples.
54
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
Fig. 2. FTIR spectrum of La3+-Mt and natural Ca2+-Mt.
3.3. La3+-montmorillonite under high pressure and high temperature (HPHT) La3+-Mt was processed at a pressure of 2.5 GPa and temperatures of 200 °C, 250 °C, 300 °C, 400 °C, 500 °C, 650 °C and 700 °C as well as at a pressure of 7.7 GPa and temperatures of 250 °C, 300 °C, 350 °C and 900 °C. EDS analysis was performed for the experiment carried out at 2.5 GPa and 700 °C for 8 h on the La3+-Mt sample. The La3+ is still
Fig. 3. In situ FTIR spectra for La3+-Mt under high pressures using the DAC system. The up arrows indicate that the pressure is being applied, while the down arrows indicate that the pressure is being released.
present in the structure at HPHT, as indicated by the high amount measured by EDS (13.20 mass%); the approximate structural formula is given by: 3þ 3þ ½Si3:13 Al0:87 O10 Al1:07 Fe0:13 Mg0:83 ðOHÞ2 La0:28 :
ð3Þ
XRD analyses were performed on all samples. Fig. 4 shows the XRD results for experiments performed at 2.5 GPa, and Fig. 5 shows the patterns for experiments performed at 7.7 GPa. All experiments were carried out for 8 h. At 2.5 GPa, the La3+-Mt structure remains stable up to 200 °C. At 250 °C, however, the (001) basal reflection shifts to a higher angle, corresponding to a basal distance of 10.5 Å; together with the other reflections, this infers that the structure has collapsed and changed to a La3+-muscovite-like structure. Although an illite-like structure was expected, the XRD reflections are in better agreement with a muscovite-like structure. For this reason, we refer to this new HPHT phase as La3+-muscovite-like from now on. Based on the structural formula calculated from the EDS results (Eq. (3)), the structure has a high layer charge; however, the nature of this phenomenon cannot be determined. As temperature increases, the La3+-muscovite-like diffraction reflections become narrower and more intense due to the more ordered structure. At 7.7 GPa, the structure of La3+-Mt is stable up to 300 °C and turns to La3+-muscovite-like at 350 °C. At 900 °C, the quartz present in the starting material becomes coesite, and the La3+-muscovite-like structure remains stable. In all experiments at both 2.5 and 7.7 GPa, the La3+-muscovite-like structure remains dioctahedral, and La3+ is part of the muscovite-like structure. When the La3+-Mt structure becomes La3+-muscovite-like, the structure loses its interlayer water, and the distance between layers drops to approximately 10.5 Å. Fig. 6 shows the detailed changes observed in the 001 reflection of the La-muscovite-like structure along with the corresponding basal distances for each HPHT condition. As
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
55
Fig. 4. XRD patterns of La3+-Mt at 2.5 GPa and different temperatures. The 001 peak of muscovite, indicated by an asterisk (*), is better specified in Fig. 7. At 200 °C, the structure remains as montmorillonite, while the patterns show muscovite-like structures at temperatures equal to and higher than 250 °C. The patterns also indicate the presence of quartz and contamination due to hBN.
Fig. 5. XRD patterns of La3+-Mt at 7.7 GPa and different temperatures. The 001 peak of La-muscovite, denoted by an asterisk (*), is better specified in Fig. 7. The structure is La3+-Mt up to a temperature of 300 °C, while it becomes La3+-muscovite-like at 350 °C and higher. The patterns also show reflections of quartz and coesite as well as low intensity reflections of hBN for the high pressure assemblage.
56
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
Fig. 6. The detailed 001 reflection of La3+-muscovite-like samples at different pressure and temperature conditions.
temperature increases, the structure becomes better ordered and the 001 reflections shift to higher angles. Ethylene glycol (EG) saturation tests were performed on the La3+muscovite-like samples obtained at 2.5 GPa/250 °C and 2.5 GPa/ 650 °C (Fig. 7) in order to analyze the behavior of their structures under these conditions. These samples were selected for EG testing because the 2.5 GPa/250 °C sample is one that is most expected to resaturate due to the lower temperature condition and the 001 reflection shape (see Fig. 6); the sample obtained at 2.5 GPa/650 °C was selected because its conditions are closest to those of the sample used in the Ca2+ adsorption experiment of the La3+-muscovite-like structure (see Section 4). None of these samples showed total or partial expansion of
the 001 distance, suggesting that there is no illite–smectite structure in these samples. However, the simple EG treatment is not sufficient to re-expand all smectite layers (Ferrage et al., 2011). EG saturation was performed as described in Bradley (1945). Fig. 8 summarizes the pressure and temperature conditions at which La3+-Mt collapses irreversibly (became a La3+-muscovite-like structure) and reversibly (remains as La3+-Mt) after the experiments. In the figure, black squares represent the experimental conditions at which the La3+-Mt structure returns to its original phase. In contrast, the black triangles represent the experimental conditions at which the La3+-Mt structure collapses to become La3+-muscovite-like and does not rehydrate afterwards. When the pressure increases, the interlayer water does leave the structure due to osmotic pressure; therefore, the structure returns to its original phase after the release of pressure. On the other hand, as temperature increases, the interlayer water has more mobility and can dehydrate, leading to an irreversible phase transition. Yet, as pressure increases, higher temperatures are necessary to generate an irreversible phase transition (Fig. 8). At these conditions, we assume that the interlayer spaces collapse and trap the lanthanum ions; we call this phase La3+-muscovite like. The phase diagram shown in Fig. 8 is similar to that of water (Wagner et al., 1994; Saitta and Datchi, 2003; Lin et al., 2004; Vega et al., 2009). To verify the stability of La inside the La3+-muscovite-like structure, we performed an ion exchange experiment in which we tried to replace La with Ca2+ in the La3+-muscovite-like structure. The experimental sample was the one produced at 2.5 GPa/700 °C/8 h because its structure exhibits good crystallinity. An XRD scan between 3° and 10° with a step of 0.02° at 4 s/step was carried out after the ion exchange experiment in order to verify the position of the first reflection (001). The diffraction pattern has two reflections: 5.73° (d(001) = 15.25 Å) and 8.62° (d(001) = 10.24 Å; Fig. 9). Therefore, two different structures are present: one that remains unchanged (La3+-muscovite-like) and another that changes into Ca 2+-Mt (15.25 Å). This result implies the presence of a mixed layer structure due to the partial readsorption of Ca 2+ by the La 3+-muscovite-like structure, turning it into the original
Fig 7. Comparison of La3+-Mt obtained at 2.5 GPa/250 °C and 2.5 GPa/650 °C saturated with ethylene glycol.
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
57
Fig. 8. Diagram based on the obtained results. Squares represent the conditions at which the structure returns to the original phase, while triangles represent the conditions at which the structure transforms.
Ca 2+ -Mt phase and releasing La3+ to the solution. As the cation exchange was performed at a pH of approximately 5, we presume that this pH is responsible for the partial cation leaching of the HPHT La3+-muscovite-like structure. Fig. 11 shows the backscattering image of this sample (Fig. 10a) along with the distributions of Ca2+ and La3+ (Fig. 10b and c, respectively). Table 2 confirms the
presence of areas enriched in Ca2+ and depleted in La3+ along with other areas enriched in La3+ and depleted in Ca2+. The approximate structural formulas calculated from the data in Table 2 are given in Eq. (4) for area 1 and Eq. (5) for area 2. 3þ 2þ 3þ Si4 O10 Al0:88 Fe0:03 Mg0:05 ðOHÞ2 Na0:65 Ca0:95 La0:08 ð4Þ
Fig. 9. XRD pattern of the La3+-muscovite-like structure processed at 2.5 GPa/700 °C/8 h and subsequently saturated with calcium. The reflection at 15.25 Å corresponds to La3+-Mt, while the reflection at 10.24 Å corresponds to the La3+-muscovite-like structure.
58
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59
Fig. 10. a) Backscattering image of the La3+-muscovite-like structure saturated with Ca2+ from the areas (1 and 2) where EDS was performed, b) Ca2+ distribution and c) La3+ distribution.
3þ 2þ 3þ ½Si3:15 Al0:85 O10 Al0:87 Fe0:249 Mg0:41 ðOHÞ2 Ca0:02 La0:43 :
ð5Þ
4. Significance to geologic and radioactive waste deposits Our experiments are applicable to two scenarios: subduction environments and radioactive waste disposal. Some amount of REE can be found in authigenic illite–smectite, as was shown by Uysal and Golding (2003); greater enrichment of heavy REE3+ is associated with a high degree of illitization, whereas less illitic clays show a higher concentration of light REE3+. During the deep burial of sediments in the subduction zone, hydrous minerals such as micas are considered to be the major carriers of near-surface H2O that is able to enter in the lower crust and mantle (Huang and Wyllie, 1973). However, smectites contain a much larger amount of water (± 15 mass%) than micas (±4.7 mass%) or even kaolinites (±13.8 mass%) and are widespread in pelagic sediments. La3+-montmorillonite (or La3+-muscovite or La3+-illite) does not occur in nature, although minor amounts of La3+ can be trapped in the interlayer spaces of montmorillonite, as described elsewhere, to share the interlayer space with other major cations such as K+, Ca2+ or Na+ (Huang and Wyllie, 1973; Uysal and Golding, 2003). When the mineral reaches pressure and temperature conditions that lead to dehydration according to the diagram above (Fig. 8), the water lost from the mineral can decrease the melting point of the surrounding rocks and promote their partial melting. Our experiments of Ca2+–La3+ replacement in the La3+-muscovite-like structure also suggest that in the presence of fluids, La3+ can be partially released from the La3+-muscovite-like structure to the surroundings, enriching the mantle in water and incompatible elements (e.g., La3+), possibly causing mantle metasomatism. However, part of the La3+ may still remain in the La3+-muscovite-like structure in deeper and hotter regions; thus, mantle incompatible element enrichment can be extended to higher pressure conditions. For radioactive waste disposal, our experiments show that La3+ (or its actinide counterpart) can be easily placed inside of La3+-Mt via cation exchange. If this mineral is exposed to high pressures and temperatures (beyond 2.5 GPa and 250 °C), it transforms into a La3+muscovite-like structure by trapping La3+. These results suggest a different approach to radioactive waste disposal where the hazardous elements are trapped in a stable crystalline structure. However, such Table 2 EDS in mass% from different areas (1 and 2) of La-muscovite saturated with Ca. The total amount sums to 100%.
Area 1 Area 2
Na2O
MgO
Al2O3
SiO2
Cl
CaO
Fe2O3
LaO3
5.30 0.00
0.57 4.17
11.75 22.16
63.56 47.74
0.00 0.04
13.96 0.33
0.68 5.02
4.18 20.54
trapping has been shown to be inefficient or partially inefficient as La3+ can be totally or partially released from La3+-Mt and La3+-muscovite-like structures, respectively, if the surrounding water leaches these mineral structures. Therefore, further studies are needed to identify a structure that completely traps lanthanum cations regardless of the surrounding environment. One of these high pressure structures could be garnet, which has a greater ability to contain REE, particularly heavy-REE3+, which better mimic the actinides. Our group is currently developing studies on this subject. 5. Conclusion Ion exchange with La3+ was achieved in Ca2+-Mt, and the structure remained very similar with the exception of the cations in the interlayer spaces. High pressures applied to the La3+-Mt structure at room temperature resulted in minor variations in the vibrational mode of the tetrahedral Si\O bond; however, this effect was reversible. When pressure and temperature were simultaneously applied to La3+-Mt, its structure remained stable up to around 2.5 GPa/250 °C and 7.7 GPa/ 350 °C; in these cases, the interlayer water had enough mobility to leave the structure, transforming La3+-Mt into a La3+-muscovite-like structure. At this point, no evidence for an illite–smectite structure was observed in our experiments. With increasing temperature, the structure became more ordered, and the HLW radionuclides could be trapped into the structure before being deposited in a multi-barrier system. Nevertheless, the La3+-muscovite-like structure in contact with an aqueous solution rich in other cations (Ca2+, for instance) would cause partial La3+ leaching; in this case, part of the structure would transform to Mt, while part of the structure would remain with the cation trapped in a La3+-muscovite-like structure. Acknowledgments We would like to thank the financial support provided by Comissão Nacional de Energia Nuclear (CNEN) from Brazil. References Alabarse, F.G., 2009. Análise da estabilidade estrutural da esmectita sob altas pressões e altas temperaturasMaster Thesis UFRGS, Porto Alegre, Brazil. Alabarse, F.G., Conceicao, R.V., Balzaretti, N.M., Schenato, F., Xavier, A.M., 2011. In-situ FTIR analyses of bentonite under high-pressure. Appl. Clay Sci. 51, 202–208. Alba, M.D., Alvero, R., Becerro, A.I., Castro, M.A., Muñoz-Paez, A., Trillo, J.M., 1997. Study of the reversibility on the local La3+ environment after thermal and drying treatments in lanthanum-exchanged smectites. Nucl. Inst. Methods Phys. Res. B 133, 34–38. Aranha, I.B., 2007. Preparação, caracterização e propriedades de argilas organofílicas. Doctorate dissertation, UFRGS, Porto Alegre, Brazil. Bergaya, F., Theng, B.K.G., Lagaly, G., 2006. Handbook of clay science. Developments in Clay Science vol. 1. Elsevier Science, Amsterdam. Bradbury, M.H., Baeysns, B., 2002. Sorption of Eu on Na- and Ca-montmorillonites: experimental investigations and modeling with cations exchange and surface complexation. Geochim. Cosmochim. Acta 66, 2325–2334. Bradley, W.F., 1945. Molecular association between montmorillonite and some polyfunctional organic liquids. J. Am. Chem. Soc. 67, 975–981.
V.F. Stefani et al. / Applied Clay Science 102 (2014) 51–59 Brandt, H., Bosbach, D., Panak, P.J., Fanghanel, T., 2007. Structural incorporation of Cm(III) in trioctahedral smectite hectorite: a time-resolved laser fluorescence spectroscopy (TRLFS) study. Geochim. Cosmochim. Acta 71, 145–154. Bundy, F.P., 1988. Ultra-high pressure apparatus. Phys. Rep. 167, 133–176. Calarge, L., Meunier, A., Formoso, M.L., 2003. A bentonite bed in Aceguá (RS, Brasil) and Melo (Uruguay) areas: a highly crystallized montmorillonite. J. S. Am. Earth Sci. 16, 187–198. Calvert, R., Prost, R., 1971. Cation migration into empty octahedral sites and surface properties of clays. Clay Clay Miner. 19, 175–186. Chapman, N.A., McKanley, I.G., Smelhe, J.A.T., 1984. The potential of natural analogues in assessing systems for deep disposal of high-level radioactive waste. Nagra Technical Report NTB, Baden, 84-41. Coppin, F., Berger, G., Bauer, A., Castet, S., Loubet, M., 2002. Sorption of lanthanides on smectite and kaolinite. Chem. Geol. 182, 57–68. Coppin, F., Castet, S., Berger, G., Loubet, M., 2003. Microscopy reversibility of Sm and Yb sorption onto smectite and kaolinite: experimental evidence. Geochim. Cosmochim. Acta 67, 2515–2527. Day, P.R., 1965. Particle fractionation and particle size analysis. In: Black, C.A. (Ed.), Methods of Soil Analysis. American Society of Agronomy, pp. 545–567. Ferrage, E., Lanson, B., Sakharov, B.A., Drits, V.A., 2005. Investigation of smectite hydration properties by modeling experimental X-ray diffraction patterns: Part I. Montmorillonite hydration properties. Am. Mineral. 90, 1358–1374. Ferrage, E., Vidal, O., Mosser-Ruck, R., Cathelineau, M., Cuadros, J., 2011. A reinvestigation of smectite illitization in experimental hydrothermal conditions: results from X-ray diffraction and transmission electron microscopy. Am. Mineral. 96, 207–223. Huang, W.L., Wyllie, P.J., 1973. Muscovite dehydration and melting in the crust and subduction oceanic sediments. Earth Planet. Sci. Lett. 18, 133–136. Kamel, G., Arai, T., Yusa, Y., Sasaki, N., Sakuramoto, Y., 1990. Estimation of illitization rate of smectite from the thermal history of Murakami Deposit, Japan. Mater. Res. Soc. Symp. Proc. 176, 657–663. Khvostantsev, L.G., 1974. High temperatures–high pressures 16, p. 165. Krauskopf, K.B., 1986. Thorium and rare-earth metals as analogs for actinide elements. Chem. Geol. 55, 323–335. Lin, J., Militzer, B., Struzhkin, V.V., Gregoryanz, E., Hemley, R.J., Mao, H., 2004. High pressure–temperature Raman measurements of H2O melting to 22 GPa and 900 K. J. Chem. Phys. 121, 8423–8427. Mookherjee, M., Redfern, S.A.T., 2002. A high-temperature Fourier transform infrared study of the interlayer and Si\O-stretching region in phengite-2M1. Clay Miner. 37, 323–336. Mozas, T., Bruque, S., Rodriguez, A., 1980. Effect of thermal treatment on lanthanide montmorillonites: dehydration. Clay Miner. 15, 421428. Piermarini, G.J., Block, S., 1975. Rev. Sci. Instrum. 46, 973. Piermarini, G.J., Block, S., Barnett, J.D., Forman, R.A., 1975. Calibration of the pressure dependence of the R1 ruby fluorescence line to 195 kbar. J. Appl. Phys. 46, 2774–2780. Pusch, R., 1998. Transport of radionuclides in smectite clay. In: Parker, A., Rae, J.E. (Eds.), Environmental Interactions of Clays — Clays and Environment. Springer.
59
Pusch, R., Karnland, O., 1988. Hydrothermal effects on montmorillonite: a preliminary study. SKB Technical Report, Stockholm 88-15. Pusch, R., Karland, O., Muurmen, A., 1989. Transport and microstructural phenomena in bentonite clay with respect to the behavior and influence of Na, Cu, and U. SKB Technical Report, Stockholm 89-34. Rabung, T.H., Pierret, M.C., Bauer, A., Geckeis, H., Bradbury, M.H., Baeysns, B., 2005. Sorption of Eu(III)/Cm(III) on Ca-montmorillonite and Na-illite. Part 1: batch sorption and time-resolved laser fluorescence spectroscopy experiments. Geochim. Cosmochim. Acta 69, 5393–5402. Saitta, A.M., Datchi, F., 2003. Structure and phase diagram of high-density water: the role of interstitial molecules. Phys. Rev. 67, 020201(R). Sherman, W.F., Stadtmuller, A.A., 1987. Experimental Techniques in High Pressure Research. John Wiley & Sons Ltd., London. Stefani, V.F., 2012. Estabilidade estrutural da esmectita dopada com lantânio sob altas pressões e altas temperaturasMaster Thesis UFRGS, Porto Alegre, Brasil. Strawn, D.G., Sparks, D.L., 1999. The use of XAFS to distinguish between inner- and outersphere lead adsorption complexes on montmorillonite. J. Colloid Interface Sci. 216, 257–269. Stumpf, T., Bauer, A., Coppin, F., Kim, J.I., 2001. Time-resolved laser fluorescence spectroscopy study of the sorption of Cm(III) onto smectite and kaolinite. Environ. Sci. Technol. 35, 3691–3694. Stumpf, T., Bauer, A., Coppin, F., Fanghanel, Th., Kim, J.I., 2002. Inner-sphere, outer-sphere and ternary surface complexes: a TRLFS study of the sorption process of Eu(III) onto smectite and kaolinite. Radiochim. Acta 90, 345–349. Stumpf, T., Hennig, C., Bauer, A., Denecke, M.A., Fanghanel, T., 2004. An EXFAS and TRLFS study of the sorption of trivalent actinides onto smectite and kaolinite. Radiochim. Acta 92, 133–138. Takahashi, Y., Minai, Y., Kimura, T., Tominaga, T., 1998. Adsorption of Eu(III) and Am(III) on kaolinite and montmorillonite in the presence of humic acid. J. Radioanal. Nucl. Chem. 234, 277–282. Tan, X., Fang, M., Wang, X., 2010. Sorption speciation of lanthanides/actinides on minerals by TRLFS, EXAFS and DFT studies: a review. Molecules 15, 8431–8468. Uysal, T., Golding, S.D., 2003. Rare earth element fractionation in authigenic illite– smectite from Late Permian clastic rocks, Bowen Basin, Australia: implications for physico-chemical environments of fluids during illitization. Chem. Geol. 193, 167–179. Vega, C., Abascal, J.L.F., Conde, M.M., Aragones, J.L., 2009. What ice can teach us about water interactions: a critical comparison of the performance of different water models. Faraday Discuss. 141, 251–276. Wagner, W., Saul, A., Pruβ, A., 1994. International equations for the pressure along the melting and along the sublimation curve of ordinary water substance. J. Phys. Chem. Ref. Data 23, 515–527.